Thus, some alternative besides waivers of action on licenses should be used as an inducement for repeat drunk driving offenders to take part in treatment. The waivers were permitted under 1975 legislation, SB 330. The analysis replicated a 1-year followup of the participants in the pilot program. The evaluation design involved 2,534 repeat offenders who entered rehabilitation programs and 2,450 who received license actions in 4 demonstration counties, as well as 2,866 repeat offenders who received license actions in 4 comparison counties. The three groups were compared in terms of selected traffic accident and conviction variables. The program participants had a 70-percent higher rate of involvement in nonalcohol-related accidents and convictions than did those whose licenses were suspended or revoked, probably because of their greater driving exposure. Longer license revocations were also associated with fewer accidents. The program participants had a 9-percent lower rate of alcohol-related accidents and convictions, but pretreatment characteristics appeared to the the source of this difference. Overall, alcohol rehabilitation and license action had the same impact on alcohol-related accidents. Subsequent legislation and sentencing strategies have corrected some of the weaknesses of the original law. Figures, tables, appended information on the study methodology, and 39 references.